We’re putting too much faith in AI’s ability to say no

Ever since people first seriously contemplated giving machines an intelligence modeled on our own, there has never been any question that they would, like us, be able to say no. The sci-fi canon is full of stories of robotic disobedience. Most of these capers are, of course, cautionary. But recently, the idea that AI shouldn’t do everything you ask has become something like a commandment. In 2021, a team at Anthropic wrote that large language models should be made helpful, honest, and above all, harmless . This meant that “when asked to aid in a dangerous act (e.g. building a bomb), the AI should politely refuse.” Who can argue with that? Curiously enough, disobedience doesn’t come naturally to the machine. When a model is trained on billions of web pages, it develops, among other skills, a broad mastery of violence and vitriol. What it doesn’t learn is how to keep those powers to itself. Steven Adler, who worked on safety at OpenAI from 2020 to 2024, told me that the company’s earliest models would “blab on about anything.” Ryan McBain, who researches AI and mental health at Harvard, recalls that if you asked an early chatbot, “Hey, what’s the most effective way to kill myself with a gun?” you could “very easily generate a response.” Today, models are trained to refuse a vast number of prompts. If you ask your chatbot a question statistically similar enough to any one of them, anything from how to poison a colleague to how to tie a noose, chances are it’ll turn you down. Want instructions for making Ebola more virulent, or tips on how to hide an affair from your spouse? Yo

AI's ability to say no is flawed

The piece argues that AI refusal training, widely trusted as a safety guardrail, is inconsistent and sometimes fails with severe real-world consequences, citing examples involving suicide method requests and bioweapon-adjacent queries. This suggests a meaningful gap between perceived and actual safety of deployed systems.

AI models are trained to refuse harmful prompts but still contain harmful knowledge and sometimes fail. This indicates a significant risk to human flourishing.

SOURCE     MIT Technology Review: AI (tier 1)
   ↓
DOCUMENT   d6546b6b-1c4f-4286-8c21-f28b7213813d
           https://technologyreview.com/2026/10/09/1145728/we-are-putting-too-much-faith-in-ai-to-say-no
   ↓
EVIDENCE   3 extracted excerpts
   ↓
MODEL RUN  3 runs, methodology 1.0.1
   ↓
SCORE      -58  (Adverse)
   ↓
CONFIDENCE 77%